DocWhisperer

A Retrieval-Augmented Generation (RAG) System for PDF Chat using Qdrant Vector Database.

Table of Contents

• Overview
• Prerequisites
• Getting Started
  • Installation
  • Usage
• Instructions for Chatting
• System Workflow
• Contributing

Overview

Imagine unlocking the hidden conversations within your dusty PDFs. DocWhisperer is your digital alchemist, transforming dry documents into vibrant dialogues. Ask anything, and DocWhisperer conjures insightful answers, pulling knowledge from the depths of your files like a magical quill. Powered by the alchemy of Qdrant's vector magic and the nimble chatbot, DocWhisperer makes them sing with meaning and ready to answer your every inquisitory whim. Step into the enchanted library of your PDFs, unleash DocWhisperer, and unlock the secrets whispered within!

Prerequisites

• Python 3
• GPU support for LLM
• Docker

Getting Started

Installation

• Run Qdrant using Docker:

  docker run -p 6333:6333 -p 6334:6334 qdrant/qdrant:latest
  

• Clone this repo:

  git clone https://github.com/Ashish-Abraham/DocWhisperer-Qdrant.git
  cd DocWhisperer-Qdrant

• Install dependencies

  pip install -r requirements.txt

Usage

 streamlit run app.py

Instructions for Chatting

Follow the prompts in the app to initiate conversations and ask questions. The chatbot will retrieve relevant information from the ingested PDFs and generate responses.

System Workflow

Contributing

• Create a branch for your changes.
• Open a pull request.